Study of realiability and validity with an application for geometry attitude scale

vm SUMMARY Doctorate Thesis STUDY of RELIABILITY AND VALIDITY WITH AN APPLICATION FOR GEOMETRY ATTITUDE SCALE Recep BÎNDAK University of Dicle Institute of Science Mathematics Program Supervisor : Assistant Prof.Cahit PESEN 2004, page 121 The aim of this study is to develop a valid and reliable a Likert type attitude scale to measure the attitudes of high school students towards geometry. This study mainly consists of three parts. The first part is the basic consepts and litterture about developing attitude scale were given. The second part is the pilot study, the Geometry Attitude Scale is improved and presented the finding about reliability and validity of this scale. The last part of present study, the prediction some personal variables on attitude toward geometry of high school students were examined. Corelations, t-tests, analiysis of variance and multiple variances techniques were used in data analyses A 46 itmes draft scale with different dimension was applied to a pilot group that consist 113 students. According to finding of item analysis 6 items are deleted. For factor analysis the remain scale was applied to 131 students. After application of item and factor analysis on the draft scale 46 items pilot form, a 25 items Likert type questionnaire was developed. To determine the reliability of the scale, the internal consistency cofficients, test retest reliability and parallel forms reliability have been estimated. Internal consistencies of the total scores were assessed by use of Cronbach alpha cofficient and item total (remainder) cerelation techniques. Cronbach alpha for the whole scale was found to be 0.942. Cronbach alpha for four sub-scale-dimension were found to be 0.908, 0.836, 0.837 and 0.702 respectively. It has a split-İA half reliability of 0.882T The test retest reliability cofficients of 0.755 and 0.847 were found for two and five week intervals, respectively. To determine the validity of the scale, structure validity, factor analiysis and similar scale validity have been estimated. Pearson product moment corelations were calculated between Geometry Attitude Scale (GAS) and geometry grades, and between GAS and math attitude scale. Positive and significants corelations cofficient were found. Factor analysis results revealed that the GAS containing 25 items loaded on four factors. Four factors were interpreted 59,26% of total original variance on scale scores. For the application of the scale Geometry attitude Scale and a questionnaire for demographic information were administred during a class hour either by the researcher herself, by teachers or by the school conselors. The data were collected from 226 female and 547 male totally 773 high school students. The final geometry attitude scale determined by means of the results of this study has been evaluated as a means of evaluation having high validity and reliability.